Intelligent monitoring method of industrial equipment based on visual recognition and big data analysis

Abstract

The present invention discloses an intelligent monitoring method for industrial equipment based on visual recognition and big data analysis, and belongs to the field of data analysis. The present invention retrieves the operating attribute data of each production equipment in the corresponding production line in the industrial production area, and performs data overall processing on each equipment number; by extracting the operating time of each equipment and dividing it into cycles, a set of real-time operating cycles of each equipment is obtained, and the operating performance status of the equipment in each cycle is analyzed respectively, and a comprehensive operating performance status analysis is performed in combination with the cycle length; the attribute data development curve of the equipment in each cycle is constructed in combination with the plane curve fitting model, and the mapping curve of the equipment operating performance status in each cycle is constructed; the attribute data development curve corresponding to each cycle of the equipment and the equipment operating performance status curve are comprehensively analyzed for the attribute performance impact correlation analysis; the real-time operating status of the equipment is evaluated in combination with the attribute performance correlation analysis results corresponding to each cycle of the equipment.

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to an intelligent monitoring method for industrial equipment based on visual recognition and big data analysis. Background Art

[0002] Intelligent monitoring of industrial equipment based on visual recognition and big data analysis adopts an integrated advanced technology to use machine vision to monitor the equipment status in real time and predict faults to ensure the safety and sustainability of equipment production;

[0003] The current production environment has increasingly strong requirements for the functionality and intelligence of production equipment. As the equipment develops intelligently, its internal structure becomes more complex and sophisticated. Therefore, traditional conventional manual safety inspections can no longer meet the equipment inspection needs in the current environment. They often require more time and human resources, and for large-scale production equipment inspection needs, they are even less efficient and less accurate. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent monitoring method for industrial equipment based on visual recognition and big data analysis to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An intelligent monitoring method for industrial equipment based on visual recognition and big data analysis, the method comprises the following steps:

[0007] S100, retrieve the operation attribute data of each production equipment in the corresponding production line in the industrial production area, and perform data overall processing on each equipment number;

[0008] S200, extracting the operation time of each device and dividing it into cycles, obtaining a set of real-time operation cycles of each device, analyzing the operation performance status of the device in each cycle respectively, and performing a comprehensive operation performance status analysis based on the cycle length;

[0009] S300, constructing the attribute data development curve of the equipment in each cycle and the mapping curve of the equipment operation performance status in each cycle respectively by combining the plane curve fitting model; analyzing the correlation between the attribute data development curve corresponding to each cycle of the equipment and the equipment operation performance status curve;

[0010] S400, evaluating the real-time operation status of the equipment based on the result of the property-performance correlation analysis of the equipment in each period.

[0011] The specific steps of S100 for retrieving the operation status data of each production equipment in the corresponding production line in the industrial production area and performing data overall processing on each equipment number are as follows:

[0012] S101. Collect equipment image data of production equipment in combination with industrial production area monitoring sensor equipment, perform three-dimensional simulation of each production equipment based on the production area image data combined with the simulation large model, construct a simulated production equipment scene, and assign regional numbers to each equipment;

[0013] S102, collecting the operating attribute data of each device through the central control sensor equipment, and transmitting the collected data to the central control management end after database integration processing.

[0014] The industrial production area monitoring sensor equipment is a monitoring camera. By networking the monitoring cameras in the production area, a monitoring network is constructed to collect images of each production equipment in the production area; the central control sensor equipment is associated with the central control management terminal, and is used to collect the operation attribute data of each equipment; the operation attribute data is the operation process data of the equipment; wherein the operation process data of the equipment includes data such as the equipment operation temperature, humidity, pressure, and rotation speed;

[0015] The S200 extracts the running time of each device and divides it into cycles to obtain a set of real-time running cycles of each device, analyzes the running performance status of the device in each cycle respectively, and performs comprehensive running performance status analysis in combination with the cycle length. The specific steps are as follows:

[0016] S201, inputting the operation attribute data collected by each device into the simulation port through the central control management terminal, performing real-time synchronization data evolution in combination with the simulation device, and obtaining the diffraction operation status data of the device corresponding to the actual scene;

[0017] S202, retrieve the actual device operation time corresponding to each simulation device, divide the operation time of each device into periods by setting a monitoring period, and construct an operation period set corresponding to each device; perform periodic analysis on the operation performance status of the device in each period in combination with the operation period set of each device; retrieve the operation attribute data of the corresponding device in the corresponding period, and analyze the operation performance of the device according to the operation attribute data of the corresponding device at the periodic time point in each period; wherein for the division of the equipment operation period, the period division is performed using the same specification time scale, and the equipment time portion whose operation time cannot be divided into a complete period is not included in the analysis step; wherein the equipment time that cannot be divided into a complete period is the excess operation time after the operation time is divided into periods, which is insufficient to divide a period, or the equipment operation time is insufficient to divide a period, then this part of time will not be included in the data analysis, and the analysis will be performed after the real-time operation time of the equipment makes up for it.

[0018] The specific steps of the equipment's operating performance analysis are as follows:

[0019] S202-1, by calling the operation cycle set divided by the current real-time operation time of the target device, locating the operation cycles in the cycle set in sequence according to the time sequence, calling the operation attribute data of the device at each time point in the corresponding cycle for the located cycle, and constructing the device operation attribute data matrix corresponding to the current located cycle;

[0020] S202-2. Analyze the performance factor scores of the equipment operation data at each time point based on the equipment operation attribute data retrieved at the corresponding time point in the positioning cycle. The calculation formula is:

[0021] ;

[0022] Among them, Pf(t∈T mn ) is the performance factor score of the operation data at the corresponding time point t in the period n divided by the corresponding device numbered m; r v is the performance correlation coefficient of the corresponding v type operation attribute data; v is the operation attribute data type number; E tv is the v-type operating attribute data at time t in the corresponding period; α(v) is the performance factor parameter corresponding to the v-type operating attribute data;

[0023] S203-3. Combine the performance factor scores of the equipment operation attribute data at each time point in the corresponding cycle to analyze the comprehensive operation performance index of the current cycle. The calculation formula is:

[0024] ;

[0025] Among them, PI(T mn ) is the comprehensive operating performance index of the cycle numbered n divided by the corresponding device numbered m; T is the length of the corresponding cycle.

[0026] The S300 combines the plane curve fitting model to construct the attribute data development curve of the equipment in each cycle and the mapping curve of the equipment operation performance status in each cycle; the specific steps of integrating the attribute data development curve corresponding to each cycle of the equipment and the equipment comprehensive operation performance status curve to perform attribute performance impact correlation analysis are as follows:

[0027] S301, using a plane curve fitting model to fit the operation attribute data of the equipment in each period and the operation data performance factor score data of the equipment at each time point in the corresponding period to a plane curve; and through plane mapping, by constructing a time axis of the same scale, mapping the fitting curves of each type of operation attribute data and the equipment operation data performance factor score data curve to the same plane coordinate system;

[0028] S302. According to the output result of the curve fitting mapping, the periodic performance impact fluctuation analysis of the corresponding type of operation attribute data is performed on the operation attribute data of each type of equipment and the performance factor score data of the corresponding equipment operation data in the coordinate system in chronological order.

[0029] Based on the corresponding period, the periodic performance impact fluctuation analysis is performed on each type of operation attribute data. The analysis is performed by analyzing the difference between the corresponding fitting curve of each type of operation attribute data and the operation data performance factor score data curve. The calculation formula is:

[0030] ;

[0031] Among them, Af v (T mn ) is the periodic performance impact fluctuation value of the corresponding v type operation attribute data; t e is the starting time point of the corresponding cycle; the average value of the periodic performance impact fluctuation value of each type of operation attribute data within the cycle is calculated according to the periodic performance impact fluctuation value obtained by analyzing the corresponding types of operation attribute data, and the calculation formula is:

[0032] ;

[0033] Among them, Af ave (T mn ) is the mean fluctuation value of the comprehensive periodic performance impact of various types of operating attribute data within the period; the difference between each operating attribute data curve and the performance factor equal division curve is integrated to obtain the difference area and analyzed in combination with the periodic scale, which is the fluctuation value of the impact of the operating attribute data on the equipment performance operation.

[0034] Based on the analysis results of the periodic performance impact fluctuation values ​​of various types of operating attribute data within the period, the abnormal fluctuation of the operating attribute data of the target equipment within each period is analyzed. The calculation formula is:

[0035] ;

[0036] Among them, API (T mn ) is the comprehensive abnormal fluctuation index of the operating attribute data in the corresponding period; PI (R) is the equipment health performance index; Af v (T mn ) max and Af v (T mn ) minIt is the maximum and minimum value of the periodic performance impact fluctuation value of the corresponding v type operation attribute data; by presetting the abnormal fluctuation index threshold of the equipment operation attribute data, when the comprehensive abnormal fluctuation index of the operation attribute data in the corresponding equipment cycle in a certain period is greater than the threshold, the current period is marked as a risk; if the comprehensive abnormal fluctuation index of the operation attribute data in the equipment cycle is less than or equal to the threshold, the equipment operates normally within the cycle.

[0037] The specific steps of evaluating the real-time operation status of the device in S400 in combination with the attribute performance correlation analysis results of the device in each period are as follows:

[0038] According to the comprehensive abnormal fluctuation index analysis results of the operation attribute data of each device in the corresponding divided operation cycle, the risk abnormal cycle marked is counted, and the abnormal cycle proportion is analyzed according to the statistical results. The calculation formula is:

[0039] ;

[0040] Among them, QT is the proportion of risk cycle; KT is the risk cycle.

[0041] According to the equipment risk cycle analysis data, by setting the risk ratio threshold, when the risk cycle ratio value is greater than or equal to the threshold, the target equipment is marked as performance abnormal; when the risk cycle ratio value is less than the threshold, the target equipment operation attribute data analysis is backed up in real time, continuous periodic data collection is performed and the above steps are repeated for abnormal analysis.

[0042] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention adopts machine vision to replace manual inspection to monitor the real-time status of production equipment, and comprehensively divides and processes the collected data through big data analysis; by discretizing the operating time of the corresponding equipment, the single-thread continuous data analysis is subjected to multi-thread discrete analysis; by analyzing the real-time performance status of the equipment, and the degree of influence of the fluctuation of each type of attribute data on the equipment performance in different periods, the abnormal performance status of the equipment is analyzed, and the risk of the equipment's operating cycle is divided by associating and analyzing the periodic abnormal fluctuation degree of each operating attribute data of the equipment with the performance index, and the equipment is abnormally marked according to the risk operation of the equipment; the present invention can realize cluster abnormality control and analysis of industrial production equipment in the field of industrial production, and realize accurate marking of equipment with abnormal performance through intelligent data association analysis and processing; it realizes accurate positioning of abnormal equipment, and effectively improves the efficiency and accuracy of manual maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0044] Figure 1 It is a schematic diagram of the steps of the intelligent monitoring method of industrial equipment based on visual recognition and big data analysis of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] See also Figure 1 , the present invention provides a technical solution:

[0047] An intelligent monitoring method for industrial equipment based on visual recognition and big data analysis, the method comprises the following steps:

[0048] S100, retrieve the operation attribute data of each production equipment in the corresponding production line in the industrial production area, and perform data overall processing on each equipment number;

[0049] S200, extracting the operation time of each device and dividing it into cycles, obtaining a set of real-time operation cycles of each device, analyzing the operation performance status of the device in each cycle respectively, and performing a comprehensive operation performance status analysis based on the cycle length;

[0050] S300, constructing the attribute data development curve of the equipment in each cycle and the mapping curve of the equipment operation performance status in each cycle respectively by combining the plane curve fitting model; analyzing the correlation between the attribute data development curve corresponding to each cycle of the equipment and the equipment operation performance status curve;

[0051] S400, evaluating the real-time operation status of the equipment based on the result of the property-performance correlation analysis of the equipment in each period.

[0052] The specific steps of S100 for retrieving the operation status data of each production equipment in the corresponding production line in the industrial production area and performing data overall processing on each equipment number are as follows:

[0053] S101. Collect equipment image data of production equipment in combination with industrial production area monitoring sensor equipment, perform three-dimensional simulation of each production equipment based on the production area image data combined with the simulation large model, construct a simulated production equipment scene, and assign regional numbers to each equipment;

[0054] S102, collecting the operating attribute data of each device through the central control sensor equipment, and transmitting the collected data to the central control management end after database integration processing.

[0055] The industrial production area monitoring sensor equipment is a monitoring camera. By networking the monitoring cameras in the production area, a monitoring network is constructed to collect images of each production equipment in the production area; the central control sensor equipment is associated with the central control management terminal and is used to collect the operation attribute data of each equipment; the operation attribute data is the operation process data of the equipment;

[0056] The S200 extracts the running time of each device and divides it into cycles to obtain a set of real-time running cycles of each device, analyzes the running performance status of the device in each cycle respectively, and performs comprehensive running performance status analysis in combination with the cycle length. The specific steps are as follows:

[0057] S201, inputting the operation attribute data collected by each device into the simulation port through the central control management terminal, performing real-time synchronization data evolution in combination with the simulation device, and obtaining the diffraction operation status data of the device corresponding to the actual scene;

[0058] S202. Retrieve the actual device operating time corresponding to each simulated device, divide the operating time of each device into periods by setting a monitoring period, and construct an operating cycle set corresponding to each device; perform periodic analysis on the operating performance status of the device in each period in combination with the operating cycle set of each device; retrieve the operating attribute data of the corresponding device in the corresponding period, and analyze the operating performance of the device according to the operating attribute data of the corresponding device at the periodic time point in each period.

[0059] The specific steps of the equipment's operating performance analysis are as follows:

[0060] S202-1, by calling the operation cycle set divided by the current real-time operation time of the target device, locating the operation cycles in the cycle set in sequence according to the time sequence, calling the operation attribute data of the device at each time point in the corresponding cycle for the located cycle, and constructing the device operation attribute data matrix corresponding to the current located cycle;

[0061] S202-2. Analyze the performance factor scores of the equipment operation data at each time point based on the equipment operation attribute data retrieved at the corresponding time point in the positioning cycle. The calculation formula is:

[0062] ;

[0063] Among them, Pf(t∈T mn) is the performance factor score of the operation data at the corresponding time point t in the period n divided by the corresponding device numbered m; r v is the performance correlation coefficient of the corresponding v type operation attribute data; v is the operation attribute data type number; E tv is the v-type operating attribute data at time t in the corresponding period; α(v) is the performance factor parameter corresponding to the v-type operating attribute data;

[0064] S203-3. Combine the performance factor scores of the equipment operation attribute data at each time point in the corresponding cycle to analyze the comprehensive operation performance index of the current cycle. The calculation formula is:

[0065] ;

[0066] Among them, PI(T mn ) is the comprehensive operating performance index of the cycle numbered n divided by the corresponding device numbered m; T is the length of the corresponding cycle.

[0067] The S300 combines the plane curve fitting model to construct the attribute data development curve of the equipment in each cycle and the mapping curve of the equipment operation performance status in each cycle; the specific steps of integrating the attribute data development curve corresponding to each cycle of the equipment and the equipment comprehensive operation performance status curve to perform attribute performance impact correlation analysis are as follows:

[0068] S301, using a plane curve fitting model to fit the operation attribute data of the equipment in each period and the operation data performance factor score data of the equipment at each time point in the corresponding period to a plane curve; and through plane mapping, by constructing a time axis of the same scale, mapping the fitting curves of each type of operation attribute data and the equipment operation data performance factor score data curve to the same plane coordinate system;

[0069] S302. According to the output result of the curve fitting mapping, the periodic performance impact fluctuation analysis of the corresponding type of operation attribute data is performed on the operation attribute data of each type of equipment and the performance factor score data of the corresponding equipment operation data in the coordinate system in chronological order.

[0070] Based on the corresponding period, the periodic performance impact fluctuation analysis is performed on each type of operation attribute data. The analysis is performed by analyzing the difference between the corresponding fitting curve of each type of operation attribute data and the operation data performance factor score data curve. The calculation formula is:

[0071] ;

[0072] Among them, Af v (T mn ) is the periodic performance impact fluctuation value of the corresponding v type operation attribute data; t eis the starting time point of the corresponding cycle; the average value of the periodic performance impact fluctuation value of each type of operation attribute data within the cycle is calculated according to the periodic performance impact fluctuation value obtained by analyzing the corresponding types of operation attribute data, and the calculation formula is:

[0073] ;

[0074] Among them, Af ave (T mn ) is the mean fluctuation value of the comprehensive periodic performance impact of various types of operating attribute data within the period.

[0075] Based on the analysis results of the periodic performance impact fluctuation values ​​of various types of operating attribute data within the period, the abnormal fluctuation of the operating attribute data of the target equipment within each period is analyzed. The calculation formula is:

[0076] ;

[0077] Among them, API (T mn ) is the comprehensive abnormal fluctuation index of the operating attribute data in the corresponding period; PI (R) is the equipment health performance index; Af v (T mn ) max and Af v (T mn ) min It is the maximum and minimum values ​​of the periodic performance impact fluctuation value of the corresponding v-type operating attribute data; by presetting the abnormal fluctuation index threshold of the equipment operating attribute data, when the comprehensive abnormal fluctuation index of the operating attribute data in the corresponding equipment cycle of a certain period is greater than the threshold, the current period is marked as a risk.

[0078] The specific steps of evaluating the real-time operation status of the device in S400 in combination with the attribute performance correlation analysis results of the device in each period are as follows:

[0079] According to the comprehensive abnormal fluctuation index analysis results of the operation attribute data of each device in the corresponding divided operation cycle, the risk abnormal cycle marked is counted, and the abnormal cycle proportion is analyzed according to the statistical results. The calculation formula is:

[0080] ;

[0081] Among them, QT is the proportion of risk cycle; KT is the risk cycle.

[0082] According to the risk cycle analysis data of the equipment, by setting the risk ratio threshold, when the risk cycle ratio value is greater than or equal to the threshold, the target equipment is marked as having performance anomalies; when the risk cycle ratio value is less than the threshold, the target equipment operation attribute data analysis is backed up in real time, continuous cycle data collection is performed, and the above steps are repeated for abnormal analysis;

[0083] In the examples:

[0084] At present, a certain industrial manufacturer performs cluster monitoring on its production equipment, and its monitoring system adopts the intelligent monitoring method of industrial equipment based on visual recognition and big data analysis of the present invention; the industrial production area monitoring sensor equipment is combined to collect equipment image data of the production equipment, and three-dimensional simulation is performed on each production equipment according to the production area image data combined with the simulation large model to construct a simulated production equipment scene, and each equipment is assigned an area number respectively; the operation attribute data of each equipment is collected through the central control sensor equipment, and the collected data is transmitted to the central control management end after being processed by database integration; the industrial production area monitoring sensor equipment is a monitoring camera, and the monitoring cameras in the production area are networked and connected to build a monitoring network to collect images of each production equipment in the production area; the central control sensor equipment is associated with the central control management end, and is used to collect the operation attribute data of each equipment; the operation attribute data is the operation process data of the equipment;

[0085] The operation attribute data collected by each device is input into the simulation port through the central control management end, and the real-time synchronous data evolution is carried out in combination with the simulation device to obtain the diffraction operation status data of the corresponding actual scene device; the actual device operation time corresponding to each simulation device is retrieved, the operation time of each device is divided into periods by setting the monitoring period, and the operation cycle set corresponding to each device is constructed; the operation performance status of the device in each period is analyzed periodically in combination with the operation cycle set of each device; the operation attribute data of the corresponding device in the corresponding period is retrieved respectively, and the operation performance of the device is analyzed according to the operation attribute data of the corresponding device at the period time point in each period;

[0086] By calling the operation cycle set divided by the current real-time operation time of the target device, the operation cycle is located in the cycle set in chronological order, and the operation attribute data of the device at each time point in the corresponding cycle is called for the positioning cycle, and the device operation attribute data matrix corresponding to the current positioning cycle is constructed; according to the device operation attribute data called at the corresponding time point in the positioning cycle, the operation data performance factor score of the device at each time point is analyzed, and the calculation formula is:

[0087] ;

[0088] Combined with the performance factor scores of the equipment operation attribute data at each time point in the corresponding cycle, the comprehensive operation performance index of the current cycle is analyzed, and its calculation formula is:

[0089] ;

[0090] The plane curve fitting model is used to fit the operation attribute data of the equipment in each period and the operation data performance factor score data of the equipment at each time point in the period; and through plane mapping, the fitting curves of each type of operation attribute data and the equipment operation data performance factor score data curve are mapped to the same plane coordinate system by constructing a time axis of the same scale; according to the output results of the curve fitting mapping, the operation attribute data of each type of equipment and the corresponding equipment operation data performance factor score data are respectively analyzed in the coordinate system according to the time sequence for the periodic performance impact fluctuation of the corresponding type of operation attribute data;

[0091] Based on the corresponding period, the periodic performance impact fluctuation analysis is performed on each type of operation attribute data. The analysis is performed by analyzing the difference between the corresponding fitting curve of each type of operation attribute data and the operation data performance factor score data curve. The calculation formula is:

[0092] ;

[0093] According to the periodic performance impact fluctuation values ​​obtained by analyzing the corresponding types of operation attribute data, the average of the periodic performance impact fluctuation values ​​of each type of operation attribute data within the period is calculated, and the calculation formula is:

[0094] ;

[0095] Based on the analysis results of the periodic performance impact fluctuation values ​​of various types of operating attribute data within the period, the abnormal fluctuation of the operating attribute data of the target equipment within each period is analyzed. The calculation formula is:

[0096] ;

[0097] By presetting the abnormal fluctuation index threshold of the equipment operation attribute data, when the comprehensive abnormal fluctuation index of the operation attribute data of the corresponding equipment cycle in a certain cycle is greater than the threshold, the current cycle is marked as a risk;

[0098] According to the comprehensive abnormal fluctuation index analysis results of the operation attribute data of each device in the corresponding divided operation cycle, the risk abnormal cycle marked is counted, and the abnormal cycle proportion is analyzed according to the statistical results. The calculation formula is:

[0099] ;

[0100] According to the equipment risk cycle analysis data, by setting the risk ratio threshold, when the risk cycle ratio value is greater than or equal to the threshold, the target equipment is marked as performance abnormal; when the risk cycle ratio value is less than the threshold, the target equipment operation attribute data analysis is backed up in real time, continuous periodic data collection is performed and the above steps are repeated for abnormal analysis.

[0101] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0102] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent monitoring method for industrial equipment based on visual recognition and big data analysis, characterized by: The method comprises the following steps: S100, retrieve the operation attribute data of each production equipment in the corresponding production line in the industrial production area, and perform data overall processing on each equipment number; S200, extracting the operation time of each device and dividing it into cycles, obtaining a set of real-time operation cycles of each device, analyzing the operation performance status of the device in each cycle respectively, and performing a comprehensive operation performance status analysis based on the cycle length; The specific steps of S200 are as follows: S201, inputting the operation attribute data collected by each device into the simulation port through the central control management terminal, performing real-time synchronization data evolution in combination with the simulation device, and obtaining the diffraction operation status data of the device corresponding to the actual scene; S202, retrieve the actual device running time corresponding to each simulation device, divide the running time of each device into periods by setting a monitoring period, and construct a set of running periods corresponding to each device; perform periodic analysis on the running performance status of the device in each period in combination with the running period set of each device; retrieve the running attribute data of the corresponding device in the corresponding period, and analyze the running performance of the device according to the running attribute data of the corresponding device at the period time point in each period; The specific steps of the equipment's operating performance analysis are as follows: S202-1, by calling the operation cycle set divided by the current real-time operation time of the target device, locating the operation cycles in the cycle set in sequence according to the time sequence, calling the operation attribute data of the device at each time point in the corresponding cycle for the located cycle, and constructing the device operation attribute data matrix corresponding to the current located cycle; S202-2. Analyze the performance factor scores of the equipment operation data at each time point based on the equipment operation attribute data retrieved at the corresponding time point in the positioning cycle. The calculation formula is: Among them, Pf(t∈T mn ) is the performance factor score of the operation data at the corresponding time point t in the period n divided by the corresponding device numbered m; r v is the performance correlation coefficient corresponding to the v type operation attribute data; v is the operation attribute data type number; E tv is the v-type operating attribute data at time t in the corresponding period; α(v) is the performance factor parameter corresponding to the v-type operating attribute data; S203-3. Combine the performance factor scores of the equipment operation attribute data at each time point in the corresponding cycle to analyze the comprehensive operation performance index of the current cycle. The calculation formula is: Among them, PI(T mn ) is the comprehensive operating performance index of the cycle numbered n divided by the corresponding device numbered m; T is the length of the corresponding cycle; S300, constructing the attribute data development curve of the equipment in each cycle and the mapping curve of the equipment operation performance status in each cycle respectively by combining the plane curve fitting model; analyzing the correlation between the attribute data development curve corresponding to each cycle of the equipment and the equipment operation performance status curve; S400, evaluating the real-time operation status of the equipment based on the result of the property-performance correlation analysis of the equipment in each period.

2. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 1 is characterized in that: The specific steps of S100 for retrieving the operation status data of each production equipment in the corresponding production line in the industrial production area and performing data overall processing on each equipment number are as follows: S101. Collect equipment image data of production equipment in combination with industrial production area monitoring sensor equipment, perform three-dimensional simulation of each production equipment based on the production area image data combined with the simulation large model, construct a simulated production equipment scene, and assign regional numbers to each equipment; S102, collecting the operating attribute data of each device through the central control sensor equipment, and transmitting the collected data to the central control management end after database integration processing.

3. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 2 is characterized in that: The industrial production area monitoring sensor equipment is a monitoring camera. By networking the monitoring cameras in the production area, a monitoring network is constructed to collect images of each production equipment in the production area; the central control sensor equipment is associated with the central control management terminal and is used to collect the operating attribute data of each device; The operation attribute data is the operation process data of the equipment.

4. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 3 is characterized in that: The S300 combines the plane curve fitting model to construct the attribute data development curve of the equipment in each cycle and the mapping curve of the equipment operation performance status in each cycle; the specific steps of integrating the attribute data development curve corresponding to each cycle of the equipment and the equipment comprehensive operation performance status curve to perform attribute performance impact correlation analysis are as follows: S301, using a plane curve fitting model to fit the operation attribute data of the equipment in each period and the operation data performance factor score data of the equipment at each time point in the corresponding period to a plane curve; and through plane mapping, by constructing a time axis of the same scale, mapping the fitting curves of each type of operation attribute data and the equipment operation data performance factor score data curve to the same plane coordinate system; S302. According to the output result of the curve fitting mapping, the periodic performance impact fluctuation analysis of the corresponding type of operation attribute data is performed on the operation attribute data of each type of equipment and the performance factor score data of the corresponding equipment operation data in the coordinate system in chronological order.

5. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 4 is characterized in that: Based on the corresponding period, the periodic performance impact fluctuation analysis is performed on each type of operation attribute data. The analysis is performed by analyzing the difference between the corresponding fitting curve of each type of operation attribute data and the operation data performance factor score data curve. The calculation formula is: Among them, Af v (T mn ) is the periodic performance impact fluctuation value of the corresponding v type operation attribute data; t e is the starting time point of the corresponding cycle; the average value of the periodic performance impact fluctuation value of each type of operation attribute data within the cycle is calculated according to the periodic performance impact fluctuation value obtained by analyzing the corresponding types of operation attribute data, and the calculation formula is: Among them, Af ave (T mn ) is the mean fluctuation value of the comprehensive periodic performance impact of various types of operating attribute data within the period.

6. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 5 is characterized in that: Based on the analysis results of the periodic performance impact fluctuation values ​​of various types of operating attribute data within the period, the abnormal fluctuation of the operating attribute data of the target equipment within each period is analyzed. The calculation formula is: Among them, API (T mn ) is the comprehensive abnormal fluctuation index of the operating attribute data in the corresponding period; PI(R) is the equipment health performance index; Af v (T mn ) max and Af v (T mn ) min It is the maximum and minimum values ​​of the periodic performance impact fluctuation value of the corresponding v-type operating attribute data; by presetting the abnormal fluctuation index threshold of the equipment operating attribute data, when the comprehensive abnormal fluctuation index of the operating attribute data in the corresponding equipment cycle of a certain period is greater than the threshold, the current period is marked as a risk.

7. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 6 is characterized in that: The specific steps of evaluating the real-time operation status of the device in S400 in combination with the attribute performance correlation analysis results of the device in each period are as follows: According to the comprehensive abnormal fluctuation index analysis results of the operation attribute data of each device in the corresponding divided operation cycle, the risk abnormal cycle marked is counted, and the abnormal cycle proportion is analyzed according to the statistical results. The calculation formula is: Among them, QT is the proportion of risk cycle; KT is the risk cycle.

8. The method for intelligent monitoring of industrial equipment based on visual recognition and big data analysis according to claim 7 is characterized in that: According to the equipment risk cycle analysis data, by setting the risk ratio threshold, when the risk cycle ratio value is greater than or equal to the threshold, the target equipment is marked as performance abnormal; when the risk cycle ratio value is less than the threshold, the target equipment operation attribute data analysis is backed up in real time, continuous periodic data collection is performed and the above steps are repeated for abnormal analysis.

Images